DESCRIPTION: The computed properties of the interatomic interactions in three complimentary examples of carbohydrate-protein complex, will provide insight into the roles played by hydrophilicity, hydrophobicity, solvation and induced fit in determining binding affinities. This approach will provide insight into the fundamental physical requirements for carbohydrate antigenicity. Experimental techniques may be used to provide ligand binding affinities, however, a computational approach must be employed for an interpretation of the experimental values in terms of discreet energetic components; e.g., electrostatic, hydrogen-bond, steric, or van der Waals contributions. Further, a computational approach enables the prediction of the effect of point mutations on binding energies. There are two categories of carbohydrate-protein interactions: those in which a compact group of sugars is recognized, such for the O-antigen of the polysaccharides from Salmonella paratyphi B (SPB), and those in which an extended epitope of 10-20 sugars is present, as in the capsular polysaccharides from group B Streptococcus (GBS) and in the interaction between polylactosamine (PL) and the S-type lectin Galectin-1. Aim 1 of this proposal is to apply the techniques of molecular dynamics (MD) simulation and free energy perturbation (FEP) to the well-characterized Salmonella-Fab system, to establish the ability of these calculations to reproduce the reported experimental geometries and interaction energies. The rigidity of the SPB ligands in this complex facilitates the accurate computation of relative binding energies. Antigenicity-enhancing interactions will be quantified. Aim 2 is to apply MD in the determination of the solution conformations of oligosaccharide fragments of PL and in the docking of these ligands to the X-ray structure of Galectin-1. The relatively simple morphology of the PL ligands facilitates a determination of the extent to which binding induces conformational changes in the receptor or ligand. Aim 3 is to determine the conformations of structurally related polysaccharides from GBS. The conformational properties of the capsular polysaccharides will be examined in order to explain the lack of mAb cross-reactivities among serotypes. Based on the results from the branched oligosaccharide (Aim1), and the linear polysaccharide (Aim 2), a model for the branched GBS capsular polysaccharides will be proposed. The role played by charge complimentarity, between the anionic polsaccharides and the protein will be assessed through FEP calculations. A better understanding of the conformational properties of the serotypes may provide a basis for rational vaccine design.